Optimizing Rain Attenuation Models for Enhanced Mobile Network Signal Performance in Lokoja, Kogi State

Rain attenuation remains a major impairment to terrestrial microwave and mobile backhaul links operating in tropical environments, where high-intensity convective rainfall events are frequent. This study investigates raininduced signal degradation in Lokoja, Kogi State, Nigeria, and evaluates the performance of the ITU-R P.530 rain attenuation prediction model under local climatic conditions. Three months of rain rate and signal attenuation data were analysed to characterize attenuation behaviour and assess model accuracy. Results indicate that while the ITU-R P.530 model performs satisfactorily under light and moderate rainfall, it systematically underestimates attenuation during intense convective rain events. Based on these findings, a location-specific optimized model was developed using a multiplicative rain-rate-dependent enhancement factor and an additive bias term, while preserving full compatibility with the ITU-R P.530 framework. The proposed optimized model reduced the root mean square error by approximately 38% and the standard error by 37% relative to the unmodified ITU-R prediction, with the most significant improvements observed at rainfall rates exceeding 30mm/hr. The enhanced prediction accuracy enables more reliable fade margin allocation and link availability estimation, thereby improving mobile network signal performance and resilience in high-rainfall tropical environments.